Enterprise Architecture Alignment Monitoring integration with a formal rule model Programme: Open University of the Netherlands, faculty of Management, Science & Technology Master Business Process Management & IT Course: IM9806 Afstudeertraject Business Process Management and IT Student: Peter Filet Identity number: 835281659 Date: June 4th, 2017 Mentor: dr. Rogier van de Wetering Examiner: prof. dr. ir. Stef Joosten Version number: 1.0 Status: Final Abstract Enterprise Architecture is an instrument that focuses on coherence between business processes, information distribution and technology infrastructure of an organization. In practice, the interrelationship between architectural aspects is not always dealt with in an integrated fashion. Enterprise Architecture frameworks are mostly informal by nature and there is a lack of knowledge and tools to support architects to check alignment in a formal manner. Due to volume and complexity of holistic enterprise spanning architecture, it is increasingly challenging for organizations to maintain overview and coherence of architectural elements. This research enables automated rule-based monitoring consistency and coherence between elements within an EA. It does so by creating an artifact that provides architects with the capability of monitoring validity within ArchiMate EA models. The models are validated against formalized rules that are specified in Relation Algebra with which coherence can be mathematically proven. The set of applied rules is plotted onto a quality framework that calculates overall alignment of an EA model. Overall alignment calculation of the researched cases show an overall alignment score varying from 47,53% to 77,27%. Every single rule violation that influences the score is identified specifically. Monitoring EA quality using formalized rules enables organizations to manage and control the process of EA change and thus contributes to Business/IT-alignment. Keywords Enterprise Architecture, Business/IT-alignment, Ampersand, Formalized rules, Relation Algebra 2 Summary Considering the last few decades, organizations are increasingly served by, and even dependent on, effective and efficient use of Information Systems and Information Technology (IS/IT). It is not surprising that within many organizations an (IT) executive-level understanding of Business/IT- alignment (also called alignment) is evolving and the topic gains priority on the executive agenda. Enterprise Architecture is an instrument that focuses on coherence between the business processes, information distribution and technology infrastructure of an organization. In practice, the interrelationship between these different architectural aspects is not always dealt with in an integrated fashion. Enterprise Architecture frameworks are mostly informal by nature and there is a lack of knowledge and tools to support architects to check alignment in a formal manner. Due to volume and complexity of holistic enterprise spanning architecture, it is increasingly challenging for organizations to maintain overview and coherence of architectural elements covering all aspects of the enterprise architecture. This statement is applicable for an architecture in the as-is state, but even more when viewed from the perspective of architecture adaptation to the to-be state. Studied literature argues that EA strengthens cohesion of business and IT. Many methods, frameworks and techniques to this end are available, but a closing theory is not found. This is consistent with what the researchers perceive in their daily practice: architects argue, discuss and suggest a lot, but provide little to none hard substantiation. The context of this research is the practice of Enterprise Architects. In order to rise above the usual practice of "arguing”, we strive for more substantiation. We also look for theory, which is applicable in practice. We state that tooling is possible that not only aids architects in automating the manual task that architects need to perform to ensure coherence, but also provides the substantiation. This research enables automated rule-based monitoring consistency and coherence between elements within an EA. It does so by creating an artifact that provides architects with the capability of monitoring validity within ArchiMate EA models. It involves EA models based on the ArchiMate modeling language. The EA models are validated against formalized rules that are specified in Relation Algebra with which coherence can be mathematically proven. The set of applied rules is plotted onto a quality framework that calculates overall alignment of an EA model. Overall alignment of the researched cases show an overall alignment score varying from 47,53% to 77,27%. Each overall alignment score is based on an underlying score for quality factors like completeness and correctness. Every single rule violation that influences the score is identified specifically. Monitoring EA quality using formalized rules enables organizations to manage and control the process of EA change and thus contributes to Business/IT-alignment. 3 Table of content Abstract ............................................................................................................................................................ 2 Summary .......................................................................................................................................................... 3 1. Introduction ............................................................................................................................................ 7 1.1. Introduction ................................................................................................................................... 7 1.2. Context ........................................................................................................................................... 8 1.3. Coherent research projects ........................................................................................................ 11 1.4. Relevance ..................................................................................................................................... 12 1.5. Problem statement ...................................................................................................................... 13 1.6. Research objective....................................................................................................................... 14 2. Literature study .................................................................................................................................... 15 2.1. Results and conclusions .............................................................................................................. 15 3. Empirical research objective ............................................................................................................... 22 4. Research method ................................................................................................................................. 23 4.1. Research strategy ........................................................................................................................ 23 4.2. Research approach ...................................................................................................................... 24 4.3. Data collection ............................................................................................................................. 27 4.3.1. EA models ............................................................................................................................ 27 4.3.2. Alignment heuristics ........................................................................................................... 28 4.4. Reliability and validity ................................................................................................................. 29 4.5. Research environment ................................................................................................................ 30 5. Research execution .............................................................................................................................. 32 5.1. Validation of alignment heuristics ............................................................................................. 32 5.2. Translating alignment heuristics to ADL .................................................................................... 33 6. Research results ................................................................................................................................... 36 7. Discussion ............................................................................................................................................. 39 7.1. Model alignment results ............................................................................................................. 39 7.2. Limitations .................................................................................................................................... 41 8. Conclusions and recommendations ................................................................................................... 42 8.1. Conclusion .................................................................................................................................... 42 8.2. Recommendations for further research .................................................................................... 43 9. Reflection .............................................................................................................................................. 44 9.1. Product ......................................................................................................................................... 44 9.2. Process .......................................................................................................................................... 44 4 9.3. Lessons learned............................................................................................................................ 45 References ..................................................................................................................................................... 46 Appendix 1 – Literature study approach, execution and detailed results ................................................ 49 Appendix 2 – Description of quality factors and attributes ....................................................................... 53 Appendix 3 – Alignment heuristics .............................................................................................................. 55 Appendix 4 – Ampersand script casus ArchiMetal ..................................................................................... 57 Appendix 5 – Ampersand script casus ArchiSurance ................................................................................. 59 Appendix 6 – Ampersand script casus Belastingdienst (NTCA) ................................................................. 61 Appendix 7 – Ampersand script casus Belastingdienst (NTCA) – meta-model ........................................ 63 Appendix 8 – Ampersand script casus Belastingdienst (NTCA) – rule TV067 .......................................... 72 Appendix 9 – Ampersand script casus Ministerie van Defensie (MOD-NL) ............................................. 73 Appendix 10 – Ampersand script casus Ministerie van Defensie (MOD-NL) – rule TV053 ..................... 75 Appendix 11 – Ampersand script casus Ministerie van Defensie (MOD-NL) – rule TV065 ..................... 76 Appendix 12 – Ampersand script casus Ministerie van Defensie (MOD-NL) – rule TV067 ..................... 78 Appendix 13 – Ampersand script casus OpenDay ...................................................................................... 80 Appendix 14 – Ampersand script casus EIRA .............................................................................................. 82 Appendix 15 – Ampersand script rule TV002 ............................................................................................. 84 Appendix 16 – Ampersand script rule TV008 ............................................................................................. 86 Appendix 17 – Ampersand script rule TV022 ............................................................................................. 88 Appendix 18 – Ampersand script rule TV023 ............................................................................................. 89 Appendix 19 – Ampersand script rule TV027 ............................................................................................. 90 Appendix 20 – Ampersand script rule TV040 ............................................................................................. 92 Appendix 21 – Ampersand script rule TV042 ............................................................................................. 93 Appendix 22 – Ampersand script rule TV043 ............................................................................................. 94 Appendix 23 – Ampersand script rule TV057 ............................................................................................. 95 Appendix 24 – Ampersand script rule TV065 ............................................................................................. 96 Appendix 25 – Rule candidates .................................................................................................................... 97 Appendix 26 – Alignment calculation example ........................................................................................ 104 Appendix 27 – Alignment measurement of case models ........................................................................ 105 5 List of tables Table 1: Effect of quality factors on overall quality (Moody & Shanks, 2003) ....................................... 17 Table 2: Similarity mapping quality factors and attributes ...................................................................... 18 Table 3: Design Science research guidelines (Hevner & Chatterjee, 2010) ............................................ 25 Table 4: Rule translation validation ............................................................................................................ 39 List of figures Figure 1: Research related to TOGAF ADM and ArchiMate (TOG, 2016) ................................................ 10 Figure 2: Architecture aspects and layers in ArchiMate 3.0 (TOG, 2016) ............................................... 10 Figure 3: Three research projects aiming to support EA coherence using formal methods ................. 12 Figure 4: Research model ............................................................................................................................ 24 Figure 5: Design Science research cycles (Hevner & Chatterjee, 2010) .................................................. 26 Figure 6: Overview of research environment ............................................................................................ 30 Figure 7: Ampersand output example ........................................................................................................ 31 Figure 8: ArchiMate visualization of rule TV065 ........................................................................................ 33 Figure 9: Total relationship .......................................................................................................................... 33 Figure 10: Simplified meta model generated by Ampersand ................................................................... 35 Figure 11: Overview of research cases and rule specifications ................................................................ 35 Figure 12: Alignment measurement result ArchiSurance......................................................................... 36 Figure 13: Overview of violations in ArchiSurance .................................................................................... 37 Figure 14: Detailed view of violations in ArchiSurance (partial) .............................................................. 37 Figure 15: ArchiMate view on Financial Application ................................................................................. 38 6 1. Introduction 1.1. Introduction Considering the last few decades, organizations are increasingly served by, and even dependent on, effective and efficient use of information systems and information technology (IS/IT). Research shows that alignment of business and IT (also called alignment) affects organizational performance (Gerow, Grover, & Thatcher, 2015; Van de Wetering, Mikalef, & Pateli, 2017b). Alignment literature generally identifies a positive relationship between the degree of alignment and business performance. However in minority, there are studies that report alignment resulting in non-existent or negative influence on business performance. This is also referred to as the “alignment paradox”, wherein a decrease in organization productivity and competitiveness has been detected. Researchers argue that alignment causes rigidity resulting to stagnation of maneuverability which is required in response to changes in the business environment (Gerow, Thatcher, & Grover, 2014). Alignment between business and IT offers value to organizations and contributes to organizational success (Castellanos & Correal, 2013; Chan & Reich, 2007; Saat, Franke, Lagerstrom, & Ekstedt, 2010). It is not surprising that within many organizations on (IT) executive-level understanding of Business/IT-alignment is evolving and the topic gains priority on the executive agenda (Gerow et al., 2015; Gerow et al., 2014; Gregor, Hart, & Martin, 2007; Pereira & Sousa, 2005). Research is available on the contribution of Enterprise Architecture (from here on referred to as EA) to Business/IT-alignment. In available literature, EA is generally considered an important instrument to contribute to Business/IT-alignment (Alaeddini & Salekfard, 2011; Castellanos & Correal, 2013; Gregor et al., 2007; Kang, Lee, & Kim, 2010; Pereira & Sousa, 2005; Sousa, Pereira, & Marques, 2004). Continuous change in demands that originate from the environment of the enterprise due to environmental hostility, customer needs or competitive stimuli, but also by the changing needs of stakeholders, creates a constant pressure on organizational performance, profitability and business continuity. Nowadays, the pace at which changes in the environment of an enterprise occur increases strongly. This forces an organization to develop dynamic capabilities to be able to adapt to changes, thus increasing complexity in both the environment and information technology (Hinkelmann et al., 2015; Steenbergen, 2011; Ullah & Lai, 2013; Van de Wetering & Bos, 2016; Van de Wetering, Mikalef, & Pateli, 2017a; Van de Wetering et al., 2017b). EA is an instrument that focuses on coherence between business processes, information distribution and technology infrastructure of an organization. In practice, the proper interrelationship between these different architectural aspects is not always dealt in an integrated fashion (Castellanos & Correal, 2013). EA, especially with (medium) large enterprises, quickly becomes large and complex. In addition, the effort to maintain EA is often carried out by a group of architects, each with their specific area of interest or specialty (Steenbergen, 2011). EA frameworks are mostly informal of nature and there is a lack of knowledge and tools to support Enterprise Architects to check this alignment in a formal manner (Castellanos & Correal, 2013; Wegmann, Balabko, Lê, Regev, & Rychkova, 2005). 7 This research aims to enable automated rule-based monitoring consistency and coherence between elements within an EA, aiding architects in achieving alignment. It involves an EA model specified in the ArchiMate modeling language. The EA model is validated against formalized rules that are specified in Relation Algebra with which coherence can be measured, monitored, and mathematically proven. The set of applied rules is plotted onto a quality framework that enables calculating overall alignment of an EA model. Overall alignment of the researched cases show an overall alignment score varying from 47,53% to 77,27%. Each overall alignment score is based on an underlying score for quality factors like completeness and correctness. Every single rule violation that influences the score is identified specifically. Monitoring EA quality using formalized rules enables organizations to manage and control the process of EA change and thus contributes to Business/IT-alignment. Chapter 1 Introduction forms the introduction of this thesis and describes the context, relevance, problem statement and objective for this research study. This research study is conducted in coherence with two other research studies in the field of EA alignment. Paragraph 1.3 specifically describes the relation between the studies. Chapter 2 describes the results of the literature study. Chapter 3 and 4 describe the goal, design and setup of the empirical part of this research. Chapter 5 describes the findings that emerged in the execution phase. An overview of the results is presented in chapter 6. Chapter 7 contains issues and views that require discussion. Chapter 8 concludes the empirical research and provides suggestions for further research and development of the artifact. Chapter 9 then describes the personal reflection of the author, reflecting on the entire research period from both a process- and product-related perspective. 1.2. Context The context of Enterprise Architecture, Business/IT-alignment and other relevant concepts to this research study is described in this paragraph. Gerow et al. (2014) describes Alignment as ‘the degree to which the needs, demands, goals, objectives, and/or structures of one component are consistent with the needs, demands, goals, objectives, and/or structures of another component (Nadler & Tushman, 1983)’. In the IT strategy literature, researchers suggest realizing the full potential of IT requires aligning some or all of four business and IT components – business strategy, IT strategy, business infrastructure and processes, and IT infrastructure and processes. Hence, IT-business strategic alignment refers to the appropriate and timely fit between two or more of these components such that management of the business and IT remain in harmony (Chan & Reich, 2007; Luftman, Papp, & Brier, 1999). On the basis of a review of the IT strategy approaches to alignment, Henderson and Venkatraman (1999) Strategic Alignment Model (SAM) describes four fundamental components of strategic choices to help organizations realize the full potential of IT. Although obtaining alignment is worth the effort due to its positive contribution to business performance, maintaining alignment provides a lasting effect. EA’s are subject to continuous change originating from the enterprise environment. In business, change is constant and misalignment between business and IT is inevitable. By adapting the EA, Business/IT-alignment can be restored. Business/IT-alignment is essential to sustain performance, effectiveness and competitiveness of organizations (Gerow et al., 2015; Gregor et al., 2007; Steenbergen, 2011; Van de Wetering & Bos, 2016). 8 The internal EA alignment is described by Sousa et al. (2004) as ‘the issue of alignment based on the coherency between elements of Business Architecture, elements of Information Architecture and elements of Application Architecture. The more elements each of these Architectures has, the richer and more complex is the concept of alignment, because more rules and heuristics need to be stated to govern the relation between these elements. So, in order to build up alignment, one must first clarify the elements of each architecture’. Achieving alignment also requires understanding of the concept of misalignment. A Business/IT- alignment model (BITAM) defines mappings between three layers of a business system: business models, business architectures and IT architectures. Misalignments in BITAM are defined as improper mappings between the layers (Chen, Kazman, & Garg, 2005). El-Telbany and Elragal (2014) define misalignment as ‘the continuous efforts, involving management and information systems, of consciously and coherently detecting and testing for the interrelation of all components of the business-IT relationship; where a change in one would instantly influence the other, contributing to the organization’s performance over time’. This research focuses on monitoring and managing the continuous changes in these interrelations. For this, we need a perceptible and automatically processable, thus formalized, form of modeling components and its interrelations. Several definitions of EA are found in contemporary scientific literature. In the context of this research, EA is described by Lankhorst (2005) as a ‘coherent whole of principles, methods, and models that are used in the design and realization of an enterprise’s organizational structure, business processes, information systems, and infrastructure.’. An EA is typically described using models, covering a holistic view of an organization. Due to complexity of an EA description, many frameworks were developed to assist in this task (Hinkelmann et al., 2015). Matthes (2011) reports more than 50 EA frameworks. Two EA frameworks are briefly mentioned here, which are widely used. In this research, we focus on the framework that is related to a modelling language that is widely used and supports automated processing. The Zachman framework (Zachman, 2002) is an enterprise ontology that provides a fundamental structure for EA. It is represented in a two-dimensional model that enables different persons to observe the same thing (i.c. an enterprise) from various perspectives. According to Hinkelmann et al. (2015), Zachman gives no advice on how the EA description should look like. Therefore the OMG1 has developed several modelling languages for EA modelling: Business Process Model and Notation- BPMN (OMG, 2011) and Case Management Model and Notation – CMMN (OMG, 2014) and the Business Motivation Model – BMM (OMG, 2015) and the UML language which is used in software development. The purpose of these graphical modelling languages is to support communication between human stakeholders. They are not intended to use for machine interpretation, which makes them unsuitable for this research study. The other well-known EA framework is TOGAF (TOG, 2011), which is composed of a set of closely interrelated architectures: Business Architecture, Information Systems (Application and Data) Architecture and Technology (IT) Architecture. TOGAF also includes a set of tools in order to enable EA teams to picture the present and future state of the architecture. TOGAF can be used in conjunction with ArchiMate (TOG, 2016). Figure 1: Research related to TOGAF ADM and ArchiMate (TOG, 2016) depicts the relation between TOGAF, ArchiMate and the focus of this research. 1 The Object Management Group (OMG) is an international, open membership, not-for-profit technology standards consortium, founded in 1989. OMG standards are driven by vendors, end-users, academic institutions and government agencies. OMG Task Forces develop enterprise integration standards for a wide range of technologies and an even wider range of industries. Source: http://www.omg.org/gettingstarted/gettingstartedindex.htm 9 Figure 1: Research related to TOGAF ADM and ArchiMate (TOG, 2016) The ArchiMate standard is based on the ISO/IEC/IEEE 42010 standard, and introduces an integrated language for describing EA’s. ArchiMate fits into the TOGAF framework as it provides concepts for creating a model that correlates to its three architecture layers. The ArchiMate modelling language is based on a formal foundation, which makes it fit for machine interpretation and thus offers possibilities for automated validation (Lankhorst, 2005). An ArchiMate model is mapped into architectural layers and aspects, as depicted in Figure 2: Architecture aspects and layers in ArchiMate 3.0. Figure 2: Architecture aspects and layers in ArchiMate 3.0 (TOG, 2016) 10